Sam Dillavou

Simons Foundation

Papers

2

Total Citations

31

H-Index

2

About

Sam Dillavou is pioneering a radical new approach to machine learning—one that does away with traditional processors altogether. His research sits at the intersection of condensed matter physics, nonlinear dynamics, and artificial intelligence, focusing on **analog learning in physical systems**. Dillavou’s central contribution is the development of **electronic contrastive local learning networks (CLLNs)**, which allow a nonlinear electronic metamaterial to “learn” directly through its own physical dynamics, without the need for digital backpropagation or a central processor. His most-cited work, “Machine learning without a processor: Emergent learning in a nonlinear analog network” (2024, 29 citations), demonstrates how these networks can perform classification tasks using only local voltage updates—a paradigm that promises orders-of-magnitude gains in speed and energy efficiency over conventional deep learning. By showing that learning can emerge from the physics of a material itself, Dillavou is laying the groundwork for **ultra-fast, fault-tolerant, and power-lean hardware accelerators**. His work has been recognized as a potential breakthrough for edge computing and neuromorphic engineering, challenging the field to rethink the very architecture of intelligent machines.

Research Focus

Key Achievements

2
H-Index
2
Papers
31
Total Citations
16
Avg Citations/Paper
🏆 Most Cited Paper
Machine learning without a processor: Emergent learning in a nonlinear analog network
29 citations · 2024
📈 Most Prolific Year: 2024 (1 Papers)
🤝 Key Collaborators: 5
🏛 Institutions: Simons Foundation

Top Papers

  1. 1
  2. 2

Key Collaborators

Contact & Links

Available for collaboration
Content generated · 13 days ago